On Thu, 15 Feb 2001, matthew green wrote:

> 
>    	there are know issues with 'float' and 'double' support on sparc64.  you
>    	should compile with -msoft-quad-float.
>    
>    -msoft-quad-float should only affect `long double' (128-bit)
>    floating port code generation.
> 
> 
> 
> yes, that is what i would understand from the switch, but programs that only
> use `double' now work with this switch.  see `iostat' or `vmstat' or `systat'.
> ie, the switch is badly named, or does more than it should (lucky!)
> 
'unsigned long' triggers compiler to use 'long double' (which might be
reasonable because there are no corresponding  machine op) and that option
forces compiler to use just 'double' instead. That what I see from
code generated. I don't know why 'hard-quad-float'd code doesn't work,
it doesn't seem wrong, I'm trying to debug it now. So if you got a chance
to check that simple test out ...

	Andrey